Biochemical Changes in Liver Associated with Kwashiorkor

Biochemical Changes in Liver Associated with

Kwashiorkor

Helen B. Burch, … , Fernando Viteri, Nevin S. Scrimshaw

J Clin Invest.

1957;

36(11)

:1579-1587.

https://doi.org/10.1172/JCI103556

.

Research Article

Find the latest version:

BIOCHEMICAL CHANGES IN LIVER ASSOCIATED WITH

KWASHIORKOR1

By HELEN B. BURCH,2 GUILLERMO

ARROYAVE,3

RUTH

SCHWARTZ,4

ANA MARIA PADILLA,3 MOISflS BflHAR,3 FERNANDO VITERI,S AND

NEVIN S. SCRIMSHAW3

(From the Department of Pharmacology, Washington University SchoolofMedicine,St. Louis, Missouri, and theInstituteofNutritionofCentral Americaand Panama (INCAP),

Guatemala City, Guatemela)

(Submitted for publication June 10,

1957;

accepted July

18, 1957)

Kwashiorkor,

the most prevalent form of

se-vere

protein

malnutrition,

is a serious

disease,

often

fatal to young children and

especially

common

from weaning to five years of age. There are

in-sufficient

data

concerning the chemical

changes

which take place

during

development

of

the

dis-ease, although clinical information is extensive.

Present knowledge about various aspects of the

disease has been reviewed

by Trowell, Davies,

and

Dean

(1),

by Brock

(2,

3)

and more

recently by

Scrimshaw

and co-workers

(4).

Its biochemical

characteristics include low levels of

serum

pro-tein, amylase,

alkaline

phosphatase,

pseudo-cholin-esterase,

cholesterol, riboflavin

and vitamin

A.

Concomitantly

the livers

are

usually

found

to be

high

in fat and low in protein.

Waterlow

(5, 6)

did

pioneering

work

on

en-zymes in liver

biopsy specimens

from

malnour-ished children and reported that hepatic

lactic

dehydrogenase

and

cytochrome oxidase

of four

children in

Gambia

remained

virtually unchanged

after

treatment

whereas pseudo-cholinesterase

was

initially

low but

more

than

doubled

on

treat-ment.

Later, Waterlow and Patrick

(7, 8)

stud-ied

levels

of eight

enzymes in liver

biopsy samples

from a

large

number of

malnourished children in

I_{This investigation was supported in part by the}

Na-tional Science Foundation, The Williams-Waterman Fund for the Combat of Dietary Diseases, the Nutrition Foundation, Inc., and National Institute of Arthritis and Metabolic Diseases of the National Institutes of Health (A-981), Public Health Service. Reported in part at

the annual meeting of the American Societyof Biological Chemists, April, 1957. INCAP Scientific Publication I-86.

2Pan

American Sanitary Bureau Consultantto INCAP from July28 to September9, 1956.

S Staff members of the Institute of Nutrition of Cen-tral America and Panama (INCAP).

4Fellow of the World Health Organization of the United Nations.

Jamaica

on

admission

to

the hospital and after

treatment.

For

cytochrome

oxidase,

lactic,

malic

and

glutamic

dehydrogenases,

succinoxidase,

DPNH-cytochrome C

reductase and transaminase

they

report

unchanged activity following

treat-ment.

The only

enzyme

found

to

be reduced in

activity in the disease

was

non-specific

cholinester-ase,

which increased

on treatment.

Extensive dietary studies in various

parts

of

the world where children suffer from kwashiorkor

reveal

inadequate intake of protein

as

well

as

low

dietary

levels of other essential nutrients

_(1,

3).

The protein is often of

poor

quality and therefore

unfavorable for the synthesis of tissue protein. If

dietary situations exist such that protein

and

other

nutrients

necessary

for the synthesis of

hepatic

enzymes are

inadequate, the levels of

some en-zymes

in the liver should reflect the lack.

Certain

hepatic

flavin

enzymes,

particularly

xanthine oxidase (9, 10), D-amino acid oxidase

(11,

12), and glycolic acid oxidase (12),

are

greatly

decreased either by riboflavin, protein

or

caloric deficiency in

rats.

Xanthine oxidase is

also lowered by the lack of a single essential amino

acid

in the diet (13, 14). The levels of these

flavo-proteins and of riboflavin might conceivably

be

related to some of the changes in metabolism

which

occur

in the livers of children with

diseases

of

malnutrition,

particularly where dietary

pro-tein and riboflavin are low.

This report gives results of biochemical

meas-urements

on

liver biopsy samples from Guatemalan

children with

kwashiorkor. Analyses

were made

for xanthine, D-amino, acid, and glycolic acid

oxi-dases,

DPNH-dehydrogenase,

malic

dehydroge-nase,

transaminase,

riboflavin, total oxidized

pyr-idine nucleotides, cholesterol, lipid and protein

in the liver as well as the

protein, cholinesterase

and amylase in

serum

and riboflavin in red blood

H. B. BURCH ET AL.

cells.

Samples

were obtained from 13

children,

in six

cases

both before

and

after

treatment.

For

comparison, levels in liver specimens

at

au-topsy from North

American

children

dying of

causes unrelated

to kwashiorkor were also

meas-ured.

To

assess

possible

postmortem

effects the

livers of normal

rats were

assayed at

various

in-tervals after

death.

EXPERIMENTAL METHODS

Sampling procedures. The children were selected as cases of typical kwashiorkor on admission to the

Gen-eral Hospital or the Hospital of the "Sociedad

Protec-tora del Nifio" of Guatemala City. If the prothrombin time ofa blood sample proved satisfactory the initial

bi-opsy sample was taken before any feeding or treatment

was given. Franseen needle biopsies were done by

as-piration through the pleural space after local anesthesia with procaine. The sample was put at once into an

ice-cold tube which remained in ice until brought to the laboratory 30 to45 minutes later.

The biopsy sample was rapidly weighed and dropped

into a small glass homogenizer, containing 15 to 20

vol-umes of 0.02 M nicotinamide at 4° C., and thoroughly homogenized. Nicotinamidewasusedtoprevent splitting of DPN by tissue DPNase. If the sample was large enough, about 5 mg. were taken for histological study. Eachsample weighed 20 mg.ontheaverage andprovided enough tissue forthe 12analyses in duplicateortriplicate. Initial biopsy samples were obtained from 10 children. Fromsixof these, biopsy samples werealsoobtained after

three to four weeks of treatment. At this time the chil-drenweregaining weight andtypicalclinical signs ofthe disease such as edema, diarrhea, and skin lesions had disappeared. Although accurate dietary control was not

possible to achieve in the wardsof the General Hospital, the children usually received a mixed diet which included

milk and supplementary vitamins. Adequate supervision ofthediet waspossible atthe Hospital of the "Sociedad Protectora del Nifto" where two children, M.A. and S.R., stayed. They received a diet rich in protein, par-ticularlymilkprotein, eating gradually increasing amounts upto5 grams of protein and 150 calories per kilo per day. Noliver samples from well-nourished Guatemalan chil-dren with no history of kwashiorkor were available for determinations of normal liver levels of chemical

sub-stances. However, to add to the group of recovered cases, biopsy samples were analyzed from three children clinically recovered from kwashiorkor, S.R., T.A. and I.T., from whom no initial samples were obtained. An-other approximation of normal hepatic levels was

ob-tained

from liver at autopsy of children

dying

of causes unrelatedtokwashiorkor in theSt. Louis Children's Hos-pital. To determine

possible

consequences of postmortem delay, livers from a series of rats were

analyzed

at in-tervalsof

0,

1,6 and 10 hours after deathfromablow on

thehead. The deadrats wereallowedtoremain at room

temperature during the specified time intervals.

Analytical procedures. Labile enzymes were measured as soon as possible after sampling. DPNH-dehydroge-nase was determined within 30 minutes after samples arrivedin the laboratory. A spectrophotometric method

was used which involved measurement of the change in optical density of DPNH at 350 my with potassium ferricyanide as the electron acceptor (12). D-amino acid and glycolic acid oxidases were measured by the method of Burch, Lowry, Combs, and Padilla (15) scaledto the useof 0.25 mg. of tissue (30

id.

incubation volumes). These methods depend upon the spectro-photometric measurement of the 3-hydrazinoquinoline derivatives of the a-keto acids formed during the action of the enzymes on D-alanineorglycolicacid, respectively, during a 30-minute incubation period. Xanthine oxidase

was measured by the rate of oxidation of 2-amino4-hydroxypteridine to fluorescent isoxanthopterin in the

presence of 5 X 10' M methylene blue (12). This method completely avoids the usual troubles from tissue blanks.

Asa rule, 10

pl.

of 1: 20 homogenate, or0.5 mg. of liver

wereusedfor eachanalysis. Riboflavin coenzymes, flavin

mononucleotide (FMN) and flavin adenine dinucleotide (FAD) were determined on the first few biopsy samples andthe autopsy samples byapublished fluorometric pro-cedure (16). On later biopsy samples only the total riboflavin was measured. Oxidized pyridine nucleotides (PN) were determined after trichloroacetic acid pre-cipitation of protein by a fluorometric procedure of Lowry, Roberts, and Kapphahn (17). Protein was de-termined calorimetrically (18).

The more stable enzymes and other substances were measured on frozen aliquots of the homogenates. For malic dehydrogenase (MDH) and aspartic-glutamic transaminase, methods developed for brain enzymes (19, 20) were adapted to liver samples. MDH was allowed

to reduce oxalacetate with DPNH, and the DPN + formed was measured byits fluorescence in alkaline solu-tion. Transaminase was allowed to act on a-keto gluta-rate and aspartate in the presence of DPNH and an

ex-cess of purified pig heart MDH. The oxalacetate pro-duced immediately oxidized DPNH to DPN+ which

was measured fluorometrically. For each determination

of MDH 0.15 y of human liver or 0.08

'y

of rat liver

were incubated in a

_12-/Al.

volume, and for each deter-mination of

transaminase

5 y of human or rat liver were incubated in

50.1d.

Cholesterol in 0.5-mg. liver samples was estimated by

thefluorometric method of Albers and Lowry (21) modi-fied by McDougal and Farmer (22) for serum. Good reproducibility and 98 to 100 per cent recovery of added cholesterol were obtained. Lipids were measured on samples of similar size. Extraction was accomplished with 3:1 alcohol-ether mixture by adding 10 volumes for

each of three extractions. The solvents were evaporated

in awater bathat 90° C. and finally the last traces were removedin a special vacuum desiccator (21). The resi-due was extracted with ether; the ether was evaporated

TABLE I

Hepaticriboflavinandflavinenzymes inkwashiorkor, initially(I)and after treatment(T)*

Total Xanthine D-amino Glycolic

DPNH-riboflavin oxidase acidoxidase acid oxidase dehydrogenase

Sex and Time mg./Kg., mM/Kg.,!hr. mM/Kg.,/hr. mM/Kg./hr. M/Kg.-/hr.

age Clinical treated

Case yrs. severity days I T I T. I T I T I T

P. G.f F-2.0 sev. 4 26 0.0 55 71 4.2

R.Ot M-5.5 mod. 10 137 1.7 134 473 17.5

M. J. F-3.0 mod. 12 138 1.6 78 20.1

L.A. F-7.0 mod. 17 108 3.5 147 362 16.3

Z. A. F-3.8 mod. 26 123 109 1.5 7.9 114 581 485 527 20.2 15.5

M. A. M-4.0 mod. 48 155 109 4.0 5.2 168 552 500 524 24.2 15.4

R. V. F-1.8 mod. 28 142 99 3.1 7.6 218 444 764 456 12.7 15.5

M.C. F-1.7 mod. 29 92 122 2.7 5.4 179 430 342 509 15.0 14.4

L. S. F-4.0 sev. 29 113 106 3.9 6.4 216 218 506 221 19.6 12.6

L. C. F-3.0 sev. 26 103 115 1.2 5.2 96 421 333 516 21.0 17.2

S. R. M-1.5 mod. 60 138 10.5 433 639 17.6

T.A. M-1.5 mod. 168 113 8.0 407 575 14.5

I. T. M-3.3 sev. 71 103 5.6 402 587 15.6

Mean 123 113 2.6 6.9 150 432 471 506 18.5 15.4

S.E. 7 4 0.4 0.6 16 34 52 42 1.2 0.5

*_{Valuesexpressed}perKg.ofliver protein.

tThese twochildrendiedinthehospital. Valuesfor P.G.weremuch lowerthanallothers and have been excluded

fromtheaverages.

off and the lipid determined by the colorimetric method

of Bragdon (23) adapted for 8 to 100 pug. of lipid in a

final volume of 0.4 ml. by Chiang, Gessert, and Lowry

(24). Trimyristin (Distillation Products Industries)

was used as the standard lipid. Quantitative recovery

of added lipid and reproducibility were obtained in pre-liminarytests madewith ratandhuman liver.

In serum, protein was measured by the gradient tube method of Lowry and Hunter (25), cholinesterase and amylase by methods of Reinhold, Tourigny, and Yonan

(26) and Smith and Roe (27). Red blood cell ribo-flavin on packed red cells was measured by a previously published method (28).

RESULTS

A.

Clinical findings

Four

of the children had severe kwashiorkor and

the

remainder were cases of moderate

severity.

All

showed typical

hair changes, skin lesions and

edema. Apathy

was a

general

characteristic,

and

all

had diarrhea.

Child

P.G. was

diagnosed

ini-tially

as

having

very

severe

and

apparently

ir-reversible

kwashiorkor and died four days after

entering

the

hospital.

B.

Hepatic

fiavin

enzymes

All enzyme and coenzyme values have been

cal-culated on

the

basis of protein. The data on

ribo-flavin

and four flavin enzymes (Table I) obtained

on

six children from whom biopsy samples were

taken

initially

and

after treatment

revealed

no

significant

rise in

riboflavin, glycolic

acid oxidase

or

DPNH-dehydrogenase.

Remarkable increases

are

apparent in

xanthine oxidase and D-amino

acid

oxidases.

These

findings

are of

special

in-terest

since,

in

the rat,

hepatic

xanthine oxidase

falls

with low dietary levels either of

protein,

calo-ries or

riboflavin, whereas DPNH-deyhdrogenase

does

not

change

in

caloric restriction

or

riboflavin

deficiency (12)

until animals

are near

death.

D-amino acid and

glycolic acid oxidases

in rat liver

also fall if

dietary

riboflavin is absent but

are

not

so

greatly affected

by

low

protein

or

calories

as

xanthine

oxidase.

Therefore,

D-amino acid

oxi-dase may not

respond

in

the same manner to

die-tary restrictions in children with kwashiorkor

as

it

does in rats.

Unusually

low levels

were

obtained

on

child

P.G. who

failed

to

respond

to treatment.

Pos-sibly the

very

low

levels of these enzymes and

other

substances

were

associated with irreversible

changes

in the

liver.

The liver

sample

of P.G.

was more

fatty

in gross appearance

than

any other.

The DPNH-dehydrogenase

is

particularly

inter-esting

as

it is

the only

really

low

value found

for

H. B. BURCH ET AL.

The values in Table I indicate

a

highly

signifi-cant

increase in hepatic

xte and

D-amino

acid oxidase levels calculated

on the

basis

of liver

protein of eight children

before and after treat-ment

of kwashiorkor.

The othet

'enzymes given

in Tables I and II did

not

increake

significantly

relative

to

liver protein.

C.

Hepatic

substances other than

_flavin

enzymes

Liver protein (Table II)

increased on

an-aver-age

of

38

per cent

(p

<

0.01) during

treatment-in the six

individuals studied

both before and

after

therapy.

At the same time total

lipid

fell to less

than one-third

the initial

value,

whereascholesterol

in

the few samples tested showed

no

significant

change.

Two

non-flavin enzymes,

transaminase,

and

malic

dehydrogenase,

chosen for

their

sig-nificance

in

relation

to

amino acid

metabolism and

the citric

acid

cycle, respectively,

weremeasuredas

controls of the flavin

enzymes.

Neither

enzyme

was

changed significantly

in

the liver

after treatment.

Pyridine nucleotides (DPN and TPN)

appear, toincrease inmostcasesupontreatment

although

the

increase is not

statistically significant.

These

nucleotides

are

destroyed

with

exceptional speed

by tissue

and red cell enzymesandmay

have been

partially split

in some

samples before

analysis

was

possible. A

change in

extent

of oxidation

of these

coenzymes would

also

affect the results. In rat

liver samples,

a

decrease in total oxidized

pyri-dine nucleotides PN occurred

when

they

were

kept in

an ice

bath

at

40

C. for

one hour without

nicotinamide and analyzed by the technique

used

on

the

biopsy samples (Table IV).

Therefore,

the levelsreported hereare

possibly

somewhat -low

but

should be comparable

before and after

treat-ment

since the specimens

were

similarly

handled,

and

enzyme

action

was

stopped by adding aliquots

to

trichloroacetic acid

atonce.

D. Protein and

enzymesinserum

A

significant increase (p

<

0.01) in

serum

pro-tein

on treatment

(Table III)

was

accompanied

by

a

significant rise (p

<

0.01)

inserum

cholines-terase relative to

protein.

The increase in serum

amylase

wasnot

significant.

If

activities

of these

enzymes are

calculated

per unit

volume

the

in-crease

become three-fold

for cholinesterase and

two-and-one-half-fold for amylase.

Thus

they

would

appear to

have

decreased

similarly

in

kwashiorkor and

to

be

synthesized

at

approxi-mately the

same rateupon recovery

from the

dis-ease.

It

is

obvious

from

the

data, however,

that

the

two enzymes behave

differently

relative to

serum

protein.

EII

Lwerconcentrationsofcertainsubstancesin

kwaskiorkor,

initially (I)andafter treatment (T)*

Protein Lipid Cholesterol Malic. dehyd. Transaminase OxidizedPN

Gm./Kg.. Gw./Kg, Gm./Kgv. M/Kg./hr. M/Kg./hr. mM/Kg.

Case I T I T I T I T I T I T

P.G.t 31 19 29 0.7

R. O. 108 995 89 85 3.8

M. J. 167 102 91

L. A. 141 107 64 2.8

Z.A. 124 150 257 23 101 132 96 76 2.8 5.9

M.A. 106 199 805 24 121 90 91 so 4.5 3.4

R.V. 87 192 833 250 28 25 176 113 91 60 3.4 4.1

M.C. 133 169 737 190 30 15 108 81 64 60 3.0 3.6

L.S. 121 167 731 219 15 24 122 105 84 56 4.1 4.6

L.C. 147 186 -115 106 69 83 3.0 4.3

S.R. 144 360 26 124 118 5.3

T.A. 172 194 16 102 66 4.1

I.T. 191 199 18 100 70 3.8

Mean 126 174 820 238 24 21 116 106 82 71 3.4 4.4

S.E. 8 6 47 19 2 2 8 6 4 7 0.3 0.3

*_{Protein per Kg.}of wettissue; others expressed per Kg. of liver protein.

TABLE III

Laeds

of

four

constitens

of

bood serm

or

cels

in

kwaskierkor

and the

efeds

of

treatment

Seam Serum

Serum cholliesterase amylase Redcell

protein unils/Gm. SuikRoeunilu/ riboflavin

Gn./1(00 d. AProw$* G.mroku ,.g./100Mi.

Case I T I T I T I T

P.G. 3.45t 275t 1.7t 24.2t

R. O. 3.77 250 13.3 10.5

L.A. 4.60 335 3.0 14.7

Z. A. 4.10 6.68 330 540 8.3 11.1 11.2 18.6

M.A. 4.36 6.91 235 890 .85.0t

7.5t

10.9 19.2

R. V. 4.50 7.68 455 935 2.9 15.0 11.0

M.C. 3.50 6.00 270 550 3.1 8.0 23.8 27.8

L. S. 3.90 6.42 400 610 20.0 12.4 24.0 39.0

L. C. 3.74 7.48 515 660 8.3 11.4 13.9 24.2

S.R. 23.2

T. A. 6.40 820 4.5

I.T. 6.75 715 14.8 26.2

Mean 4.06 6.79 3501 715§ 8.41

11.0§

15.0 25.4

S.E. 0.14 0.20 30 55 2.3 1.5 2.0 2.3

* Theunit activityof theoriginalserum asdefined byReinhold, Tourigny, and Yonan (26) multipliedby 5,000

and divided by Gm. protein per 100 ml.

t

Omittedfrom the averageasin

previous

tables.

*

ChildM. A.had hypertrophiedsalivary glands. Thesevalues havenotbeen included in theaverage.

I

Theaverage serum cholinesterase expressedasMichelUnits is 0.28 1: 0.03 initiallyand0.98:1: 0.09after treat-ment; theserum amylase expressedasSmith-RoeUnits per 100 ml. is 33 4 9and 76 F11,respectively.

Comparison

of

the increases

in

serum

protein

and

cholinesterase with that

of liver

protein (Table

II)

suggests

that

serum

cholinesterase

and

serum

protein

regenerate

simultaneously

with

liver

pro-tein

during

recovery.

The reports of Waterlow

and others

(1)

that

cholinesterase

and

protein

in

both liver

and

plasma

are

low

in kwashiorkor and

increase

greatly

on treatment

are

in

accord

with

these results. The data

on serum

amylase

aretoo

variable

to

indicate any

significant

relation

to

protein

in

liver.

Red blood cell

riboflavin increased 70

per

cent on

treatment

although the

liver riboflavin

failed

to

show such a

change.

Few data

are

available

for red

cell

riboflavin levels

in young

children, but

for four

children of similar

ages

an

average of

30

y

per 100 ml.

was

previously reported (29).

In

adults, Bessey,

Horwitt,

and Love

(30)

have

found

the red cell

level to be

a

reliable criterion

of

riboflavin

deficiency, although

in rats,

at

least,

the liver is

a more

sensitive index of mild

defi-ciency

(31). Since there

was no

other

evidence

of riboflavin

deficiency,

the

changes

may

be

sec-ondary

to

other alterations

in

the red cells.

E. Liver

specimens

at autopsy

Since

information similar to

that obtained from

livers of children with kwashiorkor was not

avail-able for

those

of normal young children,

measure-ments

were made on liver specimens from St.

Louis

autopsies.

Samples from

five

children who

seemed to be

well-nourished and who died after

periods

of one to four days in the hospital were

analyzed.

The

time before

samples were

available

varied from two to six hours after death. Levels

obtained

on

the

liver

autopsy

samples (Table IV)

are in

most instances similar to those obtained on

liver biopsy samples of the cases of kwashiorkor

after treatment.

However, xanthine oxidase and

glycolic acid oxidase activities averaged distinctly

higher in the autopsy samples (100 and

50

per

cent,

respectively). The

average

xanthine oxidase

value is five times greater than for the untreated

children with

kwashiorkor.

Since glycolic

acid

oxidase

in

kwashiorkor

did not

change

appreci-ably as the result of treatment, the low values

among

the

children concerned may reflect some

deficiency

or

disease state other

than kwashiorkor.

For

example,

this

enzyme is

a

sensitive

index

of

H. B. BURCH ET AL.

unU)'0-4N0

C14 v-4 00

v-'-4 eq*

bew~4o.4cUo

44.4CVek)0% % NO '0M 4)

.00 0N- 0

0-to_{C)Av40 00}

#-S. U)kNx )

¢Ebe m

> eu

S

Pe

S OX

be+_b

_be

tOU) eq (4a U No -4

00

WI) TV)

k-U) N

u cm

.- .

eqP

t-4J

A

o b%

0

o-o

a.- co 00

co X_ee

0

0

N04

I

-N00)01100N M 00 t~- co %O

t'-Nt'.

cU~oNMNOl+N

_%4t. m c4 c ONd

oo 00 t0 U)c

in - >m Cbs

_ o_ et

11)

0 _c 00 00 00

* 00U)Oo

~.

M*--- >

,_;:Od

04=

^4~

14c

IsU)

o64 M M M M M

1584

?I%

*e

I&

lb va

l4b

II

I.,.

00

4

,>

Cd

oq

'1'

IC

0

The rat

liver analyses

(Table IV)

are

pre-sented as a confirmation of

the

validity

of analyses

on autopsy

material.

None of the four enzymes

tested were

diminished

significantly within

six

hours of the death of

the

rat,

and

only

one,

DPNH-dehydrogenase,

was

changed

within 10 hours. The

coenzymes are not so

stable;

10 to 15 per cent

FAD was converted to

FMN

in 6 to 10

hours,

and

pyridine

nucleotides

were

either

hydrolyzed

or reduced

by

40 per

cent

in 10 hours.

Human

liver may not split FAD as rapidly as rat

liver,

however,

since FMN

was

only 10 per cent of FAD

in spite of the time lag before

obtaining

speci-mens.

It will be noted

that

total riboflavin and

all four

flavin

enzymes

are

higher

in

rat

liver

than

in

human

liver,

although transaminase is only

slightly

higher

in

the rat.

DISCUSSION

Values for liver and

serum

enzyme activities

have been

expressed

throughout

on

the

basis of

protein

because

of the well

known

changes

in

liver

and serum

protein concentrations during

treatment

of kwashiorkor.

Hepatic

flavin

enzymes

The

flavin

enzymes, xanthine and D-amino

acid

oxidase,

which were found to be low in

kwashior-kor

have

also been

shown

to

be

lowered in rat

liver

by

either

protein

or

riboflavin deficiency.

Al-though glycolic

acid oxidase is

greatly diminished

in rat

liver

by

riboflavin

deficiency, it

remains

un-changed

in human liver with treatment of

kwashi-orkor.

DPNH-dehydrogenase,

however, is not altered

in

either

the liver

of children

with

kwashiorkor

or

in

the tissues of rats

during riboflavin deficiency

or

caloric

restriction.

Wainio

and

co-workers

(32)

report that

DPNH-cytochrome C reductase

falls in

protein deficiency

in

the

rat,

but

this

has

not been confirmed in

this

laboratory

(12).

Its

failure to fall in

these livers,

except in

the

child

P.G. who died,

supports

the idea that this

is one

of the most crucial of known

flavin

enzymes and

one

of

those

most

firmly held

by liver tissue even

in various deficiency states.

This

seeming

paradox of finding no fall in

ribo-flavin,

relative to protein in liver from kwashiorkor

cases

when

some

flavin

enzymes have dropped, is

resolved

by

the

fact that these enzymes account

for

an

exceedingly small

fraction

-of

the total flavin

present

in liver.

Other

substances in the liver

It is of interest that cholesterol does not

in-crease

in

the liver in kwashiorkor according to

the values presented here. This is contrary to the

few data available on cases of kwashiorkor from

India

(33) in which much higher initial levels

were

found and a decrease

with treatment was

noted.

The

finding of virtually unchanged levels of

hepatic malic dehydrogenase and transaminase

activity

upon treatment

of

kwashiorkor

agrees

with

Waterlow and Patrick

(7). However, the

abso-lute activity found by these investigators for

trans-aminase is

only one-fifteenth of the

level reported

here.

This discrepancy may be traced to

differ-ences in

methods used. These workers also

found

rat

liver to be seven

times

lower in

transaminase

activity

than human liver.

In

the method used

here,

oxalacetate produced is immediately reduced

to

malate,

which prevents approach to

equilibrium

and consequent

slowing

of

the

reaction. Without

means for removing one of the products, larger

samples

as may

have been used by Waterlow

and

Patrick

(7)

in

their studies of rat liver would

give

lower

relative values.

It is now possible to measure a large number of

enzymes with

relative

ease in

the

amount of

ma-terial

obtained

by

liver

needle

biopsy

specimens.

Through study

of

alterations

in the enzymes, a

better

understanding

of

the

metabolic

changes

in

diseased tissues may be achieved.

Changes

in

these

functional

proteins

are

particularly

relevant

to a

deficiency

disease such

as

kwashiorkor.

SUMMARY

1. Chemical

changes in liver and serum

during

treatment

of

13 children with kwashiorkor

are

reported.

Results

of similar

analyses

on

liver

au-topsy

specimens

from

five

St. Louis children

dying

from causes unrelated to kwashiorkor are also

in-cluded, together

with

analyses

of livers

from

12

rats, to confirm the validity of studying autopsy

material.

Concomit-H. B. BURCH ET AL.

antly there

were

striking

increases relative

to

pro-tein in xanthine oxidase and

D-amino

acid oxidase.

3.

During treatment of the children with

kwashi-orkor

no

significant changes

relative to

protein

were

found in the

following

substances in liver:

riboflavin, glycolic acid oxidase,

DPNH-dehy-drogenase,

malic

dehydrogenase,

tran

nase,

oxidized

pyridine

nucleotides,

and cholesterol.

4.

The

levels of all substances measured in

the

livers of kwashiorkor cases after

treatment were

generally equal

to those

of autopsy

specimens

of

St.

Louis

children,

except

for

glycolic

acid oxidase

and xanthine oxidase which did

not increase to

the

values

found in the autopsy

samples.

5. Protein in serum increased 70 per

cent

during

treatment and relative

to

protein,

cholinesterase

increased

100 per

cent

and

amylase

30 per

cent.

The red blood

cell riboflavin

doubled.

ACKNOWLEDGMENTS

The autopsy specimens were made available through

thediligent efforts ofDr.Donald B.Strominger, towhom

the authors are greatly indebted. The authors wish to

express their appreciation to Dr. Oliver H. Lowry for

his exceedinglyhelpful advice andencouragement during

this study. They also gratefully acknowledge the as-sistance of Miss Laura Herradora and Mrs. Loty de Funes in the analyses on serum.

REFERENCES

1. Trowell, H. C., Davies, J. N. P., andDean, R. F.A.,

Kwashiorkor. London, EdwardArnold Ltd., 1954. 2. Brock,J. F., Surveyof the worldsituationon

kwashi-orkor in Conference on Nutritional Factors and Liver Diseases,Annals of the New York Academy

of Sciences. New York, New York Academy of Sciences, 1954,vol. 57,p. 696.

3. Brock, J. F., Nutrition in Annual Review of

Bio-chemistry. Stanford, Annual Reviews, 1955, vol. 24, p. 523.

4. Scrimshaw,N.S.,Behar, M.,Arroyave,G., Viteri, F.,

and Tejada, C., Characteristics of kwashiorkor (Sindrome pluricarencial delainfancia).

Federa-tionProc., 1956, 15, 977.

5. Waterlow,J. C., Observations ontheactivity of some enzymes in the human liver. West Indian M. J.,

1951, 1, 41.

6. Waterlow, J. C., Enzyme activity in human liver in

Liver Injury, Transactions of the Eleventh

Con-ference, F. W. Hoffbauer, Ed. New York, Josiah

MacyJr. Foundation, 1953, p. 72.

7. Waterlow, J. C., and Patrick, S. J., Enzyme activity in fatty livers in human infants in Conference on

Nutritional Factors and Liver Diseases, Annals of

the New York Academy of Sciences. New York, New YorkAcademy of Sciences, 1954, vol.

57,

p. 750.

8. Protein Malnutrition; Proceedings of a conference in Jamaica (1953) sponsored jointly by the Food and Agriculture Organization of the United Na-tions, the World Health Organization and the Josiah Macy Jr. Foundation, J. C. Waterlow, Ed. Cambridge, Cambridge Univ. Press, 1955, p. 16. 9. Axelrod, A. E., and Elvehjem, C. A., The xanine

oxidasecontentof ratliver in riboflavindeficiency. J. Biol. Chem., 1941, 140, 725.

10. Richert, D. A., and Westerfeld, W. W., Some

inter-relations of dietary protein,molybdenum, riboflavin

and calories on liver and intestinal xanthine oxi-dase. Proc. Soc. Exper. Biol. & Med., 1953, 83,

726.

11. Axelrod, A. E., Sober, H. A., and Elvehjem, C. A., The D-amino acid oxidase content of rat tissues

in riboflavin deficiency. J. Biol. Chem., 1940, 134, 749.

12. Burch, H. B., Lowry, 0. H., Padilla, A. M., and

Combs, A. M., Effects of riboflavin deficiency and

realimentation on flavin enzymes of tissues. J. Biol. Chem., 1956, 223, 29.

13. Williams,J.N.,Jr.,andElvehjem, C. A., Therelation of amino acid availability in dietary protein to liver enzyme activity. J. Biol. Chem., 1949, 181,

559.

14. Williams, J. N., Jr., andElvehjem, C. A., The effects of tryptophan deficiency upon enzyme activity in the rat. J. Biol. Chem., 1950, 183, 539.

15. Burch, H. B., Lowry, 0. H., Combs, A. M., and

Padilla, A. M., A sensitive methodfor measuring

amino acid oxidases. Apr., 1956, Abstracts, Div. of Biol. Chem., Am. Chem. Soc., 129th meeting, Dallas.

16. Burch,H. B., Fluorimetricassayof FAD, FMN,and

riboflavin inMethods in Enzymology, S. P.

Colo-wick and N. 0. Kaplan, Eds. New York, Aca-demic Press Inc., 1957,vol. III, p. 960.

17. Lowry, 0. H., Roberts, N. R., and Kapphahn, J. I., The fluorometric measurement of pyridine nucleo-tides. J. Biol. Chem., 1957, 224, 1047.

18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and

Randall,R.J., Protein measurement with the Folin

phenol reagent. J. Biol. Chem., 1951, 193, 265. 19. Lowry, 0. H., Roberts, N. R., and Lewis, C., The

quantitative histochemistry of the retina. _J. Biol.

Chem., 1956, 220, 879.

20. Lowry, 0. H., Roberts, N. R., andChang, M. L. W.,

Theanalysis of single cells. J. Biol. Chem., 1956,

222, 97.

21. Albers, R. W., and Lowry, 0. H., Fluorometric de-terminationof 0.1 to 10 micrograms ofcholesterol. Anal. Chem., 1955, 27, 1829.

22. McDougal, D. B., Jr., and Farmer, H. S., A fluoro-metric methodfor total serum cholesterol. J. Lab. & Clin. Med., 1957, 50, 485.

23. Bragdon, J. H., Colorimetric determinatonof blood bhatll d itre i ,a qantities of blood lipids. J. BioL Chem, 1951, 10, 513. serumandcelils. J,- Biol. Chem., 1948, 175, 457. 24. Chiang, S. P., Gessert, C. F., and Lowry, 0. H., 29.

Snyderman,

S. E., Retron, K. C., Burch, H. B.,

Personal

ommunication.

Lowry,

0.-H

_{7Bems_,}

9.-A.

Guy,

LXP., andHolt, 25. Lowry, 0. H., and Hunter, T.H.,Thedeterminatier L E.,

jr

e minimum riboflavin requirement of

of serum proteinconcentration with a gradient tube. theinfant. Nutrition, 1949, 39, 219.

J. Biol. Chem., 1945, 159, 465.i30.BeseT, 0. A., Horwltt M. -E and Love, R H., 26. Reinhold, J. G., Tourigny, L. G., and Ynain, V.-L., Dietary deprivation of riboflavin and blood ribo-Measurementof serumcholinesterase activity by a flavin levels in man. J. Nutrition, 1956, 58, 367. photometric indicator methoc Tb' eer with a 31.

Lo*y,

O. H.,-Bioddkhical evidence of nutritional study of the influence of sex and race. Am. J. status. Physiol. Rev., 1952, 32, 431.

Clin.

Path.,

1953,- 23 645. 32.;

Wainio,

W. W.,BEicel,-

B.,

Ziche, H.

J.,

Person, P., 27. Smith, B. W., and Roe,J.-H., A photometric method Istes, F. Lo.

_*nd

l J B 0xidatve enzymes for the determination of

B-ainylase

in blood and oftheliver in

protein

3ion

_J.

Nuttion,

1953, urine, with use df the

starch-iodine

color. J. Biol. 49,

46S.

Chem.,-1949, 179, 53. 33.

Ramalingaswami,

V., Srichari, S., and Tulpule, 28. Burch, H. B., Bessey, 0. A., and Lowry, 0. H., P. G., Hepatic cholesterol content In nutritional